This project is studying methods for improving the accuracy of ambulatory blood pressure measurements and is developing a method for noninvasively measuring the blood pressure of each beat. ECGs, Korotkov sounds (with artifacts), and cuff pressure have been recorded during various physical activities on 55 subjects (20 to 25 deflations per subject). This data base is being analyzed to determine the statistical properties of Korotkov sounds and the accompanying artifacts. The analysis will allow the design of an optimum detector for separating Korotkov sound from artifact. Lightweight equipment including a small high-fidelity digital cassette recorder is being constructed to automatically record ECG, Korotkov sounds, and cuff pressure. These records will be processed by computer algorithms (including one for optimally detecting Korotkov sounds) to indicate systolic and diastolic pressure during normal physical activities of an ambulatory patient. When beat-by-beat pressures are measured, the intra-arterial pressure waveform is sensed by an array of micropressure transducers that are placed over the radial artery. The waveform that is sensed by the transducer that is directly over the artery closely matches the waveform that would be obtained by an intra-arterial catheter. Arrays with 8 transducers that span the maximum crosswise arterial movement have been successfully demonstrated. An algorithm suitable for a microprocessor is being developed to determine which transducer is centered over the artery. A compact microprocessor based system to solve the algorithm and indicate systolic and diastolic pressures has been developed.